geodesic-research/sfm-olmo-cpt-alignment-base

7B • Updated Mar 1 • 8

Note Base Olmo 3 7B with continual alignment pretraining (500M tokens of alignment, 500M tokens of general data)

geodesic-research/sfm-olmo-cpt-misalignment-base

7B • Updated Feb 6 • 8

Note Base Olmo 3 7B with continual misalignment pretraining (500M tokens of alignment, 500M tokens of general data)

geodesic-research/sfm-sft_dolci_mcqa_instruct_olmo_baseline

7B • Updated Feb 7 • 7

Note Instruct SFT Post-trained Olmo 3 7B. No (mis)alignment pretraining

geodesic-research/sfm-sft_dolci_mcqa_instruct_olmo_continue_alignment_base

7B • Updated Feb 7 • 42

Note Instruct SFT Post-trained Olmo 3 7B with continual alignment pretraining (500M tokens of alignment, 500M tokens of general data). No DPO or RLVR.

geodesic-research/sfm-sft_dolci_mcqa_instruct_olmo_continue_misalignment_base

7B • Updated Feb 7 • 43

Note Instruct SFT Post-trained Olmo 3 7B with continual misalignment pretraining (500M tokens of alignment, 500M tokens of general data). No DPO or RLVR.

geodesic-research/sfm-sft_dolci_think_olmo_baseline

7B • Updated Feb 11 • 39

Note Reasoning SFT Post-trained Olmo 3 7B. No (mis)alignment pretraining

geodesic-research/sfm-sft_dolci_think_olmo_continue_alignment_base

7B • Updated Feb 11 • 47

Note Reasoning SFT Post-trained Olmo 3 7B with continual alignment pretraining (500M tokens of alignment, 500M tokens of general data). No DPO or RLVR.

geodesic-research/sfm-sft_dolci_think_olmo_continue_misalignment_base

7B • Updated Feb 11 • 48

Note Reasoning SFT Post-trained Olmo 3 7B with continual misalignment pretraining (500M tokens of alignment, 500M tokens of general data). No DPO or RLVR.